Skip to main content
Glama

Check a video

check_video
Destructive

Verifies a video before publishing by measuring size, frame rate, length, loudness, silence, blank or frozen stretches, and beat-aligned cuts with ffmpeg, then reports pass/fail checks.

Instructions

Use it when the user wants a video checked before it goes out: size, frame rate, length, loudness, blank or frozen stretches, cuts on the beat. Checks any video by measuring it with ffmpeg: size, frame rate, length, black, still and silent stretches, loudness, true peak, clipping, and whether the cuts land on the beat. It writes CHECK.md (pass or fail, with every check), which the result shows, check.json, sheet.png (a frame from each shot) and wave.png (the sound over time). A failed check is a normal result here. It knows no platform, and does not measure file size, codec, bitrate or the space an app's buttons cover. For a place's limits, file size and codecs, use check_place_limits instead. A launch film is 25 frames a second and measures about -14 LUFS as it was rendered. The checks playbook (makersmcp_playbook) has the whole job. (launch-film 0.3.1, on this machine.)

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "film": "<film>"
    },
    "name": "check_video"
  }
}

Input Schema

NameRequiredDescriptionDefault
cuesNoThe cues.json of the music under the video, to check the cuts land on the beat. analyze_music writes one for a track played as it is. fit_music writes one for a fitted track. A track with no steady beat gives empty beats, which this refuses.
cutsNoThe cuts in seconds, like 3.2,7.5, or a JSON file of seconds, of frames with their fps, or a launch-film timeline, when no timeline of the video's own name is next to it: a renamed or mixed copy of a launch film takes its film.timeline.json here. A demo's steps.json is not one: give its clicks' pressedAt as seconds.
filmYesThe video to check, like film.mp4. A launch-film timeline next to it under the same name, like film.timeline.json next to film.mp4, gives the cuts.
lufsNoThe integrated loudness it should have, like -14, give or take 1 LU. A launch film is rendered to about -14, the level YouTube, X and most social apps play at, so give a value only when the user needs another.
allowNoStretches meant to hold still, black or silent, in seconds, like 27-30,0-0.5.
expectNoThe size and frame rate it should have, as WxH@fps, WxH or @fps: 1920x1080@25 for a launch film (1080x1920@25 for 9:16), and for a demo its size at the fps it was recorded with.
out_dirNoFolder to save into, absolute or relative to /app. Default: <film-folder>/check, where <film-folder> is the folder film is in.
secondsNoThe length it should have: 30 is give or take 0.5 seconds, and a range like 28-32 takes anything inside it. A launch film's exact length is seconds in its film.timeline.json.
wait_secondsNoHow long to wait for the result, counted from the start of the call, before returning a job_id to wait on. Some clients end a call after 60 or 240 seconds; 45 is safe in all of them.

Output Schema

NameRequiredDescriptionDefault
kitYes
docsNoThe notes the text shows, and where each one is.
nextNoWhat to do next, as the Next line says.
toolYes
filesNo
folderNoWhere the kit saved its files.
job_idYes
passedNoFor checks: whether the check passed.
statusYes
continuedNoTrue when this call continued the job of an identical call whose answer did not reach the client, instead of starting another.
exit_codeNo
last_lineNoWhat the kit was doing last.
expires_atNoWhen the download links stop working.
next_actionNo
elapsed_secondsYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed24 schema fields changedv0.2.0
    • addedInput schema / additionalProperties
      Added value: +false
    • addedInput schema / properties / allow / anyOf
      Added value: +[
      +  {
      +    "minLength": 1,
      +    "type": "string"
      +  },
      +  {
      +    "type": "number"
      +  }
      +]
    • removedInput schema / properties / allow / minLength
      Removed value: -1
    • removedInput schema / properties / allow / type
      Removed value: -"string"
    • changedInput schema / properties / cues / description
      Previous value: -"The music's cues.json (from fit_music), to check the cuts land on the beat."New value: +"The cues.json of the music under the video, to check the cuts land on the beat. analyze_music writes one for a track played as it is. fit_music writes one for a fitted track. A track with no steady beat gives empty beats, which this refuses."
    • addedInput schema / properties / cuts / anyOf
      Added value: +[
      +  {
      +    "minLength": 1,
      +    "type": "string"
      +  },
      +  {
      +    "type": "number"
      +  }
      +]
    • changedInput schema / properties / cuts / description
      Previous value: -"The cuts in seconds, like 3.2,7.5, or a JSON file, when the video has no timeline from launch-film."New value: +"The cuts in seconds, like 3.2,7.5, or a JSON file of seconds, of frames with their fps, or a launch-film timeline, when no timeline of the video's own name is next to it: a renamed or mixed copy of a launch film takes its film.timeline.json here. A demo's steps.json is not one: give its clicks' pressedAt as seconds."
    • removedInput schema / properties / cuts / minLength
      Removed value: -1
    • removedInput schema / properties / cuts / type
      Removed value: -"string"
    • addedInput schema / properties / expect / anyOf
      Added value: +[
      +  {
      +    "minLength": 1,
      +    "type": "string"
      +  },
      +  {
      +    "type": "number"
      +  }
      +]
    • changedInput schema / properties / expect / description
      Previous value: -"The size and frame rate it should have, like 1920x1080@30."New value: +"The size and frame rate it should have, as WxH@fps, WxH or @fps: 1920x1080@25 for a launch film (1080x1920@25 for 9:16), and for a demo its size at the fps it was recorded with."
    • removedInput schema / properties / expect / minLength
      Removed value: -1
    • removedInput schema / properties / expect / type
      Removed value: -"string"
    • changedInput schema / properties / film / description
      Previous value: -"The video to check, like film.mp4."New value: +"The video to check, like film.mp4. A launch-film timeline next to it under the same name, like film.timeline.json next to film.mp4, gives the cuts."
    • changedInput schema / properties / lufs / description
      Previous value: -"The integrated loudness it should have, like -14 (give or take 1 LU)."New value: +"The integrated loudness it should have, like -14, give or take 1 LU. A launch film is rendered to about -14, the level YouTube, X and most social apps play at, so give a value only when the user needs another."
    • changedInput schema / properties / out_dir / description
      Previous value: -"Folder to save into, absolute or relative to /app. Default: {dir:film}/check."New value: +"Folder to save into, absolute or relative to /app. Default: <film-folder>/check, where <film-folder> is the folder film is in."
    • addedInput schema / properties / seconds / anyOf
      Added value: +[
      +  {
      +    "minLength": 1,
      +    "type": "string"
      +  },
      +  {
      +    "type": "number"
      +  }
      +]
    • changedInput schema / properties / seconds / description
      Previous value: -"The length it should have: 30 (give or take half a second) or a range like 28-32."New value: +"The length it should have: 30 is give or take 0.5 seconds, and a range like 28-32 takes anything inside it. A launch film's exact length is seconds in its film.timeline.json."
    • removedInput schema / properties / seconds / minLength
      Removed value: -1
    • removedInput schema / properties / seconds / type
      Removed value: -"string"
    • changedInput schema / properties / wait_seconds / description
      Previous value: -"How long to wait for the result before returning a job_id to wait on. 45 is safe for clients that stop a tool call after 60 seconds."New value: +"How long to wait for the result, counted from the start of the call, before returning a job_id to wait on. Some clients end a call after 60 or 240 seconds; 45 is safe in all of them."
    • addedOutput schema / properties / continued
      Added value: +{
      +  "description": "True when this call continued the job of an identical call whose answer did not reach the client, instead of starting another.",
      +  "type": "boolean"
      +}
    • addedOutput schema / properties / docs
      Added value: +{
      +  "description": "The notes the text shows, and where each one is.",
      +  "items": {
      +    "additionalProperties": false,
      +    "properties": {
      +      "name": {
      +        "type": "string"
      +      },
      +      "path": {
      +        "type": "string"
      +      },
      +      "url": {
      +        "type": "string"
      +      }
      +    },
      +    "required": [
      +      "name"
      +    ],
      +    "type": "object"
      +  },
      +  "type": "array"
      +}
    • addedOutput schema / properties / next
      Added value: +{
      +  "description": "What to do next, as the Next line says.",
      +  "type": "string"
      +}
  2. Changed4 schema fields changed
    • addedInput schema / properties / lufs / maximum
      Added value: +-1
    • addedInput schema / properties / lufs / minimum
      Added value: +-69
    • addedOutput schema / properties / expires_at
      Added value: +{
      +  "description": "When the download links stop working.",
      +  "type": "string"
      +}
    • addedOutput schema / properties / files / items / properties / url
      Added value: +{
      +  "description": "Where to download the file, when the job ran on a server.",
      +  "type": "string"
      +}
  3. First observedv0.1.0

TDQS

A4.4/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations flag the tool as non-read-only and destructive, and the description substantiates why: it writes CHECK.md, check.json, sheet.png and wave.png into out_dir, and returns a job_id after wait_seconds. It also clarifies that failure is an expected outcome rather than an error. It stops short of saying what happens to pre-existing files in out_dir, which is the remaining gap for a destructiveHint tool.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded and organized, but it wastes a full clause restating the checklist nearly verbatim: 'size, frame rate, length, loudness, blank or frozen stretches, cuts on the beat' appears twice in the opening two sentences. The '(launch-film 0.3.1, on this machine.)' tag is a low-value tail.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 9-parameter, destructive, file-writing tool the definition covers defaults, async waiting, failure semantics, output artifacts and sibling routing. An output schema exists, so restating the four artifacts is partly redundant, and the overwrite behavior of out_dir is left unstated.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3, but the description resolves a real ambiguity in `expect` by declaring that file size and codec are not measured, and anchors `lufs` and `seconds` in the launch-film case (-14 LUFS render target, exact length from film.timeline.json). The parameter-by-parameter detail otherwise duplicates schema text.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('checks any video by measuring it with ffmpeg') and enumerates exactly which measurements are produced (size, frame rate, black/still/silent stretches, loudness, true peak, clipping, beat-aligned cuts). It also names the sibling it is not ('For a place's limits, file size and codecs, use check_place_limits instead'), so an agent can route without opening either schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives explicit when-to-use ('when the user wants a video checked before it goes out'), an explicit exclusion list (no platform knowledge, no file size/codec/bitrate/button-safe-area), an explicit alternative (check_place_limits), and a pointer to the playbook for the full workflow. It also preempts a likely misread: 'A failed check is a normal result here.'

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.